Lupeol and its esters : NMR , powder XRD data and in vitro evaluation of cancer cell growth

1Departamento de Produtos Farmacêuticos, Faculdade de Farmácia, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brasil, 2Departamento de Química, Centro de Ciências Exatas e Naturais,Universidade Estadual de Ponta Grossa, Ponta Grossa, Paraná, Brasil,3Departamento de Química, Instituto de Ciências Exatas, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brasil, 4Centro Pluridisciplinar de Pesquisas Químicas, Biológicas e Agrícolas, Universidade Estadual de Campinas, Paulínia, São Paulo, Brasil, 5Faculdade de Ciências Farmacêuticas, Universidade Estadual de Campinas, Campinas, São Paulo, Brasil, 6Departamento de Farmácia, Escola de Farmácia, Universidade Federal de Ouro Preto, Ouro Preto, Minas Gerais, Brazil


INTRODUCTION
Despite the efforts to develop new strategies of cancer prevention and therapy (Galmarini, Galmarini, Galmarini, 2012), cancers still represent a worldwide problem of public health.According to the Word Health Organization (Ferlay et al., 2012), around 14 million of the new cancer cases occurred in 2012 (57% of this total in less developed regions).Among the strategies to treat cancer, cancer chemotherapeutic agents represent crucial tools basically aiming to eliminate or at least inhibit tumor cell growth (Galmarini, Galmarini, Galmarini, 2012;Chabner, Roberts, 2005).Considering all antitumor chemotherapeutic arsenal approved between 1940s and 2014, 49% (85 chemical entities) were natural products per si or directly derived from them (Newman, Cragg, 2016).
The aim of this work was the semi-synthesis of eight lupeol esters, from which five (5 to 9) are new ones.The compounds were characterized by Fourier transform infrared (FT-IR), nuclear magnetic resonance ( 1 H and 13 C NMR) spectroscopy, CHN analysis and powder X-ray diffractometry (XRD).Moreover, lupeol and the eight lupeol esters were evaluated regarding their antiproliferative in vitro potential against a human cell lines panel.

Synthesis and identification of lupeol esters
Lupeol (1) was isolated from hexane branch extract of M. salicifolia through phytochemical processes as described in the literature (Magalhães et al., 2011).The esters 2 to 9 were obtained reacting 1 with an adequate carboxylic acid and the DIC/DMAP reagents (Figure 1), with yields ranging from 86 to 96%.The 1 H and 13 C NMR chemical shift assignments of compound 1 (S3) were in accordance with the spectral data published by Shahlaei and coworkers (2013).
The structures of lupeol esters (2 to 9) were confirmed due to the disappearance of signal at δ C 71.0, in the 13 C NMR spectra ( S4 to S11), corresponding to carbon 3 bonds in the hydroxyl group, the presence of signal at ~ δ C 80.0 associated to C-O-C, together the of the signal at ~ δ C 171.0 (C=O).The signal associated to C=O group to compound 9 appeared at δ C 166.10 due to the influence of 3',4'-dimethoxybenzoate group (Mahato, Kundu, 1994).The physical chemical data (IR, 1 H and 13 C NMR and CHN analysis) of compounds 1 to 9 are described below, as well as the amount obtained (mmol) and percent yield for the esters 2 to 9.
For the XRD experiments, the material was homogenously spread over the sample holder under spinning to prevent preferred orientation and minimize rugosity effects over the exposed surface.The small amount submitted to the XRD, few milligrams, was composed essentially of polycrystalline material.Single crystals were not identified or isolated from the synthetic material.So, detailed crystallographic data were provided only for the isolated lupeol.For lupeol (1), the angles are 90.0 due to the special positions on tetragonal P43 space group symmetric restrictions.Other details of refinements and X-ray diffraction experimental data are summarized in Table I.Due to the small amount of esters (2 to 9), all fittings were obtained at P-1 space group that safely allowed us to index all peaks.After extracting and fitting, all peaks in space group P-1 were searched for more symmetric space group based on the Bragg systematic absences.More symmetric space groups were achieved for compounds 1, 7 and 8.The remaining ones have not shown any symmetric description based on the systematic Bragg absences.The powder XRD data of lupeol esters 2 to 9 (Table I) were consistent with the 13 C NMR data of each one indicating the tendency of the compounds to be in the crystalline state.The XRD experiment was considered as an excellent tool to determine the structure of lupeol and its esters in solid state.

Cell proliferation assays
All the compounds were tested for proliferation of human cancer cells.Doxorubicin (anthracycline) used as positive control is a chemotherapy drug that decreases or stops the growth of cancer cells.The activity of doxorubicin involves blocking the enzyme called topo isomerase 2 that cancer cells need to replicate and grow.Lupeol was inactive (GI 50 >250 µg/mL) in the experimental condition while lupeol esters 2-4 and 7-9 showed a cytostatic effect on colorectal adenocarcinoma (HT-29) and chronic myelogenous leukemia (K-562) cell lines (Table II).
The introduction of a long alkyl side chain (2) in lupeol resulted in a cytostatic effect on the colorectal adenocarcinoma (HT29) cell line (GI 50 = 97.81µg/mL).This effect increased by reducing the length of the alkyl chain, from C16 (2) to C12 (4), resulting in the best effect (GI 50 = 1.74 µg/mL).However, the continuous reduction in the chain length (from C10 to C4, compounds 5-8) afforded inactive compounds (GI 50 >250 µg/mL).Moreover, lupeol palmitate (3) showed a selective growth inhibition effect on erythromyeloblastoid leukemia (K-562) cells in a concentration-dependent way (Figure S2).Previous studies have shown that the palmitic acid is active against leukemic cells (Harada et al., 2002).Probably, the activity observed for lupeol palmitate is due to the fatty moiety.Thus, compound 3 could be considered as a prototype for the development of new anticancer drugs to be used in leukemia treatment.For the aryl lupeol ester (9) it was seen a quite similar cytostatic effect (GI 50 =0.95µg/mL) to that observed for ester 4. On the other hand, the side chain length seemed not to be as influent against chronic myelogenous leukemia (K-562) cells as for antiproliferative activity against HT-29.In this regard, only compounds 5 (with ten carbon in side chain) and 6 (with eight carbons in side chain) were not able to inhibit K-562 cell proliferation up to 250 µg/ mL (Figure S2, Table II).The esters 4 and 9 showed a selective cytostatic effect with low GI 50 values (Figure S2), therefore, these compounds represent a promising prototype for the development of new anticancer drugs.

CONCLUSION
The esters 2 to 9 were obtained using lupeol, an adequate carboxylic acid and DIC/DMAP reagents, with yields ranging from 86 to 96%.The esters 5 to 9 were new compounds.The XDR method was an excellent tool to determine the structure of lupeol and its esters in solid state.Lupeol esters 3, 4 and 9 showed a selective cytostatic effect with low GI 50 values, representing a promising prototype for the development of new anticancer drugs.

General experimental procedures
Melting points (mp) (uncorrected) were determined using a Mettler FP 80 HT apparatus. 13C NMR spectra were obtained on a Bruker Avance DRX 400 or on Bruker DPX 200 spectrometers.The sample was dissolved in CDCl 3 and TMS was used as internal standard (δ C = 0).IR spectra were recorded on a FITR-Perkin-Elmer, Spectrum One SN 74759 spectrophotometer.Powder X-ray diffraction (XRD) data were collected in an XRD-7000 diffractometer (Shimadzu, Japan) under 40 kV, 30 mA, using Cu Kα (λ = 1.54056Å) equipped with a polycapillary focusing optics under parallel geometry coupled with a graphite monochromator, scanned over an angular range of 4−70° (2θ) with a step size of 0.01° (2θ) and a time constant of 5 s.step −1 .The sample holder was submitted to a spinning of 30 cycles per minute to minimize rugosity effects and to reduce any eventual preferred orientation.The lattice parameters were extracted and fitted by Rietveld fitting analysis.CHN analyses were performed in a Perkin Elmer, Series II, CHNS/O Analyzer.Classical chromatographic column (CC) was carried out using silica gel 60 (Merck,.TLC was obtained using pre-coated silica gel plates, and the detection was visualized by spraying the plates with solution (1:1) of vanillin (ethanol 1 % solution w/v) in perchloric acid (3% aqueous solution v/v), in accordance with Wagner and Bladt (1996).

Plant material
Maytenus salicifolia Reissek (Celastraceae) was collected at 'Serra de Ouro Branco', a mountain located in the Ouro Branco City region, Minas Gerais (MG) state, Brazil.The plant was identified by Dr. Rita Maria Carvalho-Okano, Botanist of the Universidade Federal de Viçosa, MG, Brazil.A voucher specimen of M. salicifolia was deposited (Nº.OUPR-18094) at the Herbarium José Badini of the Universidade Federal de Ouro Preto, MG, Brazil.

Isolation of lupeol and synthesis of esters
The isolation of lupeol was reported by Magalhães and coworkers (2011).For the esters synthesis, the following sequence was carried out for the reactions: to 1.0 mmol of lupeol (1), x mmol of carboxylic acid and y mmol of 4-(dimethylamino)pyridine (DMAP) in 7.0 mL of dry dichloromethane were added (Table I).After cooling down to 0 °C and under constant magnetic stirring, z mmol of N,N'-diisopropylcarbodiimide (DIC) was carefully added.Then, the reaction mixture was maintained under magnetic stirring, at room temperature, for 2 to 48 hours depending on the carboxylic acid used as reagent.The reaction time was monitored by TLC using CHCl 3 -MeOH (9.5:0.5) as mobile phase.The reaction conditions of carboxylic acid with lupeol [1, (1.0 mmol)] and DIC/DMAP to obtain the lupeol esters 2 to 9 (Figure 1) are presented in Table SI.
At the end of the reaction, the dichloromethane was recovered in a rotator evaporator and the residual material obtained from each esterification reaction was purified by chromatographic column eluted with CHCl 3 .The lupeol esters 2 to 4 (Figure 1) were obtained as a white waxy material while lupeol esters 5 to 9 were obtained as a white amorphous solid.

Characterization of compounds
The structure of lupeol and its synthesized esters were initially characterized by IR, NMR ( 1 H, 13 C) and CHN data.The spectral results were carefully compared with data available in the literature (Mahato, Kundu, 1994).Then, the structure of each compound was fitted through powder XRD.Thus, compound 1 (or ester 2 to 9) was reduced to a very fine powder and deposited as a film suspension in a Zero Field Sample Holder (ZFSH) composed by polished SiC in a 3° angle mount to reduce background contributions for the X-ray diffraction experiment.The powder indexing tool used was Conograph (Oishi-Tomiyasu, 2012) for cell and space group determination, followed by Pareto optimization and Rietveld with energies fitting of the structure.The peaks were searched and fitted with a David-Voight approximation peak profile, performing both modified Pawley and Rietveld with energy refinements to optimize powder diffraction parameters and crystal structure, so that the best possible agreement between simulated and experimental powder pattern was achieved.Lattice parameters were expressed in nanometers (nm) and the angles in degrees (0).The fitted uncertainties were listed with the significant figures obtained.Due to the small amount of material (lupeol esters), all fittings were obtained at P-1 space group, which allowed us to index all peaks safely.To search for more symmetric Space Group occurrences more natural extracted material would be necessary to increase low intensity peaks that may help search for more symmetry in all diffractograms.After extracting and fitting all peaks in space group P-1, a search was performed for more symmetric space groups based on the Bragg systematic absences.Details of refinements and experimental data of X-ray diffraction are summarized in Table I.

Sample preparation
Aliquots (5.0 mg) of lupeol and its esters 2 to 9 were initially diluted in DMSO (50 µL) followed by the addition of 950 µL of RPMI 1640/FBS 5% (working solution).The solutions were then diluted in RPMI 1640/FBS 5% in order to obtain the final concentrations.DMSO final concentrations (≤ 0.25%) in culture medium did not affect the cell viability.

TABLE I -
Lupeol (1)and its esters (2 to 9) lattice parameters obtained by Rietveld fitting of the powder X-ray diffraction